While there are several imaging technologies that offer spatial resolution in studying the brain, none can provide the temporal resolution of electrophysiological techniques. Conversely, the latter cannot compete with the former in terms of anatomical detail. Therefore, there is a natural desire to integrate different modalities of data and images for the study of brain function and the diagnosis of disorders. The Human Motor Control Section of the NINDS has been using EEG, MEG, transcranial magnetic stimulation (TMS), MRI, and PET to study voluntary movement for several years. A method for integrating several or all of these approaches was considered highly desirable. A method was developed for mapping equivalent dipole sources, computed from EEGs and MEGs, into MRI and PET. Later, this method was extended to include scalp maps obtained by TMS of the brain. The precision of the mapping method that we developed is 3 mm, which is adequate for integration of the various signals.